An introduction to scientific software tools and parallel algorithms

This course provides a comprehensive, hands-on introduction to the tools and workflows used in modern scientific software development and an applied introduction to parallel algorithms and performance analysis. Topics include:

• Unix shell and HPC usage (ssh, tmux, Slurm job scheduler)

• Version control using Git (local and collaborative workflows)

• Automated documentation (Doxygen for C/C++, Sphinx/MyST for Python and general docs)

• Unit and functional testing (Google Test for C++, pytest for Python)

• Containerization and reproducibility (Docker, Apptainer/Singularity)

• Continuous Integration and automation (GitHub Actions)

• Basics of parallel computing: speedup, efficiency, Amdahl’s and Gustafson’s laws

Learning outcomes

By the end of the course, students will be able to:

• Operate comfortably in a Linux command-line environment and write basic Bash scripts.

• Submit and monitor jobs on a Slurm-based cluster (we provide a Dockerized simulator for exercises).

• Use Git to manage projects, branch, merge, resolve conflicts, and collaborate via pull requests.

• Produce documentation with Doxygen and Sphinx/MyST, including API references and tutorial pages.

• Write unit and functional tests for C++ and Python code and integrate them into CI.

• Build and run Docker and Apptainer containers to package software environments for reproducibility.

• Set up simple CI workflows (GitHub Actions) to run tests and build artifacts automatically.

• Measure simple parallel programs, compute speedup and efficiency, and interpret results using Amdahl’s and Gustafson’s laws.

Practical Information

Semester: 1st semester (Nov 2025 – Jan 2026)

Credits: 6 ECTS (30 hours)

Language: Italian/English

Exam: project